The number of Americans with two or more chronic conditions will increase from 57 million to 81 million people over the next 20 years. It is not clear what constitutes optimal health outcomes for persons with multiple morbidities ('complex patients'), how to attain these outcomes, or how to measure this attainment. Both complex patients themselves and expert recommendations emphasize the need for patient-centered care including continuity of relationships with clinicians and coordination of care. However, we do not know which patient-level factors affect care outcomes, whether continuity of care should be a primary component of care for complex patients, and if so, which types of clinicians should establish those continuity relationships. To address these questions we will build on previous investigations that have concentrated on small pieces of the overall process of care for complex patients and start 'putting the pieces together'to inform practical change in our healthcare system. We hypothesize that a) subjective, patient-level factors such as financial constraints and perceived disease burden, are important in achieving desired health outcomes for complex patients;b) the effect of these patient- level factors is mediated by interpersonal continuity of care, and c) continuity of care need not be only with the primary care physician (PCP). Rather, effective interpersonal continuity of care may be provided by care managers or specialty physicians as well. In order to test these hypotheses we will assess a range of health outcomes as a function of subjective factors important to complex patients in a study population of approximately 900 adults age 65+ with 3 or more chronic medical conditions. We will combine collection of subjective data by a 2-phase survey with 2-year follow up, with substantial electronic data on patient and clinical variables including detailed encounter and continuity data. We will then develop mediational models to assess health outcomes as a function of continuity of care and patient-level factors. Although clearly a function of effective systems and policies, the focus of medical care is the patient. It will be virtually impossible to implement effective systems- or policy-level interventions to improve care without an understanding of what matters to complex patients. Coordinated, team-based care has been described as a crucial component of efficient and effective care of this population. This implies that effective continuity of care need not be focused solely on the patient-PCP relationship, but could occur with other clinicians as well. However, such 'multidimensional'continuity has never been studied with regard to patient needs or health outcomes. Results from this investigation will clarify the benefits and mechanism of a broadly recommended, but unproven process of care for complex patients.